Data lake

There's the event streams.

And then finally, there's data lakes.

And data lakes are another really, really interesting technology that kind of... comes up out of Hadoop and has gone through these transformations.

But they're the place where you put your – they're now called lake houses.

They're the place where you put very large pools of data that are just out there as references for things like reports to see if you're complying with regulations, stuff like that.

This is an incredibly capable database.

A database like ClickHouse is the thing that allows you to analyze these piles of data that are coming, served by Kafka, coming off these sort of vast armies of software services.

And then finally, Data Lakes is the place you can exhaust the data once you just want to put it out and sort of as a reference where data scientists and other people who need direct access the large amounts of data can just access it and use it in creative ways.

But I'll tell you one thing which we have done, which I didn't mention.

At Data Lake, we invested very heavily this year in putting all the data, not only the customer data and the metadata, but also internal data and data from all the systems across the company into this lake house.

And that's kind of our source of truth right now.

So it doesn't matter if you're in a go-to-market team or a product team or engineering team, or if you have our product like people using, all the data is going into lake house and we are building agents on top of that for every team right now.

But basically what this means is it can, it changes the pricing of your ingestion.

It spins up Data Lake for you because there's been, There's been issues with capacity, compute capacity for data lake.

This is getting rid of that in effect because it's using Fabric as the way to mechanism to be able to sort of do some extra querying against it, I believe.

So this is definitely sort of a big step and it's available for everyone that's got E5 or E7 licensing.

Well, say the standard pricing in the US is, I'm going to say $5.40, I think it is, per gig per day.

I think this drops down to $2.50 or $2.30 or something like that.

So in effect, half price on ingestion for your sort of analytic storage, plus you've got the data lake for all that, you know, all the data.

And really, this is the steer that agentic AI or AI agents that are either built or that come with XDR, they have the data lake to help them to be able to make decisions and things like that based on your environment, because you're able to store a lot more data for it to then have context and things like that.

So they're investing in OpenAI or Claude or Gemini to do a bunch of work and search across their company.

Then they're investing in a data lake to organize all their data.

And then they come to you and they're like, I'm surprised because 93% of my projects are stuck in pilot.

The model is cool and my data is all organized.

I was kind of waiting for the data angel to come and save me.

but now the model didn't save me and the data angel didn't save me.

So what do I do? And then my answer back to them is, well, we have to go back to data lakes, capture a record of everything that you've done in the past, right? Great job.

You've organized all your promotions.

And now you have all these vendors coming out going,

You know, Data Lake and then

Data Lake House and like, you know what I mean? Now you're getting vendor sprawl in that world.

Absolutely.

And like, well, what are we supposed to do? Like, do we get rid of this stuff? Like you can't get, you

We have been achieved embryo static validations.

And our next steps gonna be, we would like to build Data Lake, in which it will serve as a single startup group for whole organizations and connect critical and operational data for AIs and data analytics.

And so that opportunity about artificial intelligence can be exciting, but sometimes healthcare organizations aren't sure where they might start with that journey.

What do you think is the biggest opportunity within your facility for utilizing artificial intelligence?

We have advantage because we have clinical data integrated in track care system.

However, the challenge is like how we connect those clinical data with operational data from other systems.

So that's why we plan to build our data lake as a single sort of route.

That makes a lot of sense.

There's a lot of misunderstanding there.

So in this case, what these guys are trying to do, they are trying to align with the snowflake because data lake, data warehouse should not be the term.

I'm using it loosely.

But...

This is going to be your real data lake lake that you are trying to consolidate all of those systems.

Sometimes 09 could be a feed to that data lake or data lake could feed to 09 depending upon your interaction model in terms of what you may be trying to accomplish.

So again, we have seen these kind of synergies pretty much everywhere because Snowflake, Databricks, they all are going after pretty much every single enterprise software vendor out there.

So overall, architecturally speaking, and from the CXO's perspective, this is great because you are going to get probably a pretty big connection.